Umbrella Trial Design Calculator
Designs a biomarker-stratified umbrella trial — one disease, multiple biomarker-defined sub-studies, each testing a different targeted treatment against a shared control arm. Supports binary, continuous, and survival endpoints with frequentist or Bayesian analysis, and estimates operating characteristics by simulation.
How it works
Patients are stratified into biomarker-defined sub-studies, each of which compares its treatment to the common control. Sharing one control arm across sub-studies improves efficiency relative to running separate trials. You can apply a multiplicity correction (e.g. Bonferroni or Holm) across sub-studies, and the tool reports per-sub-study power and error rates.
When to use it
- You are studying one disease with several biomarker-defined treatments and want a shared-control design.
- You want to reuse a common control arm to reduce the number of control participants.
- You need per-sub-study power and a defined approach to multiplicity across sub-studies.
Assumptions & limitations
- The efficiency of a shared control depends on the control group being comparable across sub-studies; using it validly generally requires concurrently randomized, comparable control participants.
- Each sub-study should be adequately powered on its own — the shared control improves efficiency but does not remove the need for sufficient per-sub-study sample sizes.
- Whether and how to apply multiplicity correction across sub-studies is a design choice; treating different targeted treatments as distinct clinical questions may not require strong family-wise control, while related comparisons may.
- Reported operating characteristics are simulation estimates for the modeled scenario, not guarantees for a realized trial.
For the full methodology, derivation, and worked examples, read the complete guide.